Abstract: A critical challenge facing medicine is the development of more effective and less toxic systemic therapies for cancer patients. A new paradigm for the development of systemic cancer therapeutics has emerged over the last decade with the move away from broadly cytotoxic agents to targeted therapies. This has been enabled by the identification of specific alterations that drive oncogenesis in a wide variety of tumor types and the development of small molecules or antibodies that specifically target these oncogenic drivers. In some genetically defined subsets of cancer patients whose tumors harbor an oncogenic driver (e.g. mutant EGFR, ALK gene fusions, mutant B-RAF), pharmacologic oncogene inhibition has become standard of care. Despite dramatic clinical successes achieved with inhibitors of driver kinases that promote tumor growth, responses in patients are rarely complete and also vary in duration. Strategies to enhance initial responses and to prevent or overcome resistance to inhibition of oncogenic drivers are needed. A significant obstacle to the optimal use of targeted cancer therapies is the lack of a coherent strategy to develop appropriate combinations that will enhance response rates and prevent or overcome resistance. Defining systematically rational companion targets whose combined inhibition can achieve maximal therapeutic efficacy and improve the survival of cancer patients is a major and seemingly insurmountable challenge facing oncology. Fortunately, emerging functional genomics technologies provide an unprecedented opportunity to accelerate the design of effective targeted cancer polytherapies. The goal of this proposal is to create an intellectual foundation and experimental platform that will optimize the personalized treatment of cancer patients and improve their survival using as a model system lung cancer, the leading cause of cancer mortality in the US. We propose a conceptually and technically innovative, systematic, and interdisciplinary approach to efficiently design effective cocktails of targeted cancer therapies that synergistically destroy human lung cancers by acting against essential oncogenic networks in tumor cells. Our unified approach integrates complementary tools including cancer genomics, unbiased double RNA interference loss of function genetic screens, systems network analyses, clinical therapeutics, and prospectively acquired human clinical data. Our approach will allow us to discover unexpected mechanisms that regulate the dependence of a tumor on a driver oncogene that promotes its growth and to identify rare synergistic gene interactions and signaling networks in oncogene- driven tumors that allow them to evade treatment with an oncogene inhibitor. The payoff of these studies will be both an improved understanding, at the basic level, of the functions of human oncogenes and, at the translational level, a new model to accelerate the development of effective rational combinations of targeted therapies at unprecedented scale. If successful, we envision that this approach will lead to effective cancer polytherapies for molecularly defined subsets of lung cancer patients and can be applied to cancers broadly. Public Health Relevance: Cancer is a major public health problem because it is the second most common cause of death in the US. Personalized treatments that specifically target proteins that drive cancer growth in an individual patient are leading to improved responses in some cancer patients but an overwhelming obstacle to the success of such targeted therapy is treatment resistance. To overcome this challenge, we propose a conceptually and technically innovative, systematic, and interdisciplinary approach that is transformative because it will allow us to efficiently design effective cocktails of targeted cancer therapies that synergistically destroy human lung (and other) cancers and, thus, will have a major impact on the ability of personalized treatments to increase the survival of, and potentially even cure, patients with lung and other lethal cancers.